Molecular and cellular mediators of innate and adaptive immune responses play a major role in both the initiation and the progression of stroke. They also participate in the induction of tolerance to ischemia by sublethal preconditioning stresses. We have investigated such mediators in the three abovementoned facets of stroke. Inhibition of tumor necrosis factor(TNF) with TNF-binding protein reduces brain infarct volume in middle cerebral artery occlusion (MCAO) models in the rat and mouse. In addition, TNF-binding protein attenuates the progressive impairment of microvascular perfusion that occurs during the early hours of focal brain ischemia. These findings implicate TNF as a mediator of progressive brain damage during acute stroke. Lipopolysaccharide (LPS) pretreatment has been demonstrated to induce tolerance to focal brain ischemia in the MCAO model in spontaneously hypertensive rats (SHR). TNF binding protein blocks this tolerance implicating TNF as a mediator of this state of preserved homeostasis under stress. This form of tolerance also reduces the degree of microcirculatory perfusion impairment in brain. Preconditioning with TNF by intracisternal injection of TNF induces tolerance to ischemia in the Balb/C mouse. In vitro models comprising cellular elements of brain have been established in order to examine the mechanisms involved in the observed in vivo tolerance to ischemia of the brain pretreated either by TNF or by oxygen/glucose deprivation (OGD). Pretreatment of primary neuronal cultures with short hypoxia (15 minutes) 24 hours prior to 60 minutes of hypoxia, protected neurons against hypoxia (number of dead cells was 9.4% versus 35% in non-pretreated cultures). Pretreatment with TNF (50 ng/ml) also protected cortical neurons against 60 minutes of hypoxia. TNF-preconditioning can induce tolerance to subsequent OGD in brain microvessel endothelial cells, astrocytes and cortical neurons. In cortical astrocytes from 2-3 day old Sprague-Dawley rats, preconditioning with TNF-alpha to produce tolerance does not inhibit I-kappaB proteolysis, nuclear translocation of NF-kappaB or binding of the p65 subunit of NFkB to its consensus site on DNA. It does, however, prevent p65 phosphorylation and consequently disrupts the association of the coactivator protein, p300/CBP, with that subunit. The result is that expression of proinflammatory genes such as ICAM-1 is inhibited, but expression of cytoprotective genes such as manganese superoxide dismutase continues unabated. In bedside to bench studies, mouse anti-rat ICAM-1 antibody induced an inflammatory state in preclinical models of ischemic stroke that included activation of complement (C3a desArginine), granulocytes (CD11b up-regulation), and endothelium (E- and P-selectin expression). Serial administration of the antibody sensitized rats to produce anti-mouse antibodies and augmented infarct size in a focal brain ischemia model. Similar responses to the mouse anti-human ICAM-1 monoclonal antibody (a non-humanized antibody), Enlimomab, may have contributed to the adverse outcomes of the Enlimomab acute stroke trial. Immunological tolerization to E-selectin, an adhesion molecule that only becomes expressed on endothelium when a vessel segment becomes activated, has been shown to profoundly reduce ischemic strokes and to eliminate hemorrhages that otherwise occur spontaneously in spontaneously hypertensive and genetically stroke-prone rats (SHR-SP). E-selectin tolerization also can reduce infarct size after MCAO in SHR-SP and this cytoprotection can be adoptively transferred by splenocytes from tolerized animals indicating that the protection is cell-mediated. This novel approach to stroke prevention is being translated into clinical trials involving the secondary preventon of stroke. GMP recombinant murine and human E-selectin has been produced with a baculovirus expression vector platform and is being used in preclinical toxicology and immunotoxicology studies in mice and non-human primates as a prerequisite for obtaining an IND. Additional studies have shown that E-selectin tolerization markedly inhibits white matter damage in models of vascular dementia and experimental autoimmune encephalomyelitis. E-selectin tolerization also suppresses delayed vasospasm in a subarachnoid hemorrhage model. A bedside to bench proposal has supported the study of E-selectin tolerization in a model of atherosclerosis where the tolerization suppresses lipid deposition in the aortic arches of APOE null mice on a Western diet. These studies are conducted in collaboration with the NINDS Neuroimmunology Branch and NHLBI personnel. Current work follows a critical path to Phase I testing and includes testing E-selectin tolerization with acetylsalicylic acid in preclinical stroke models since the mucosal tolerization with E-selectin approach will be used as an "add on" rather than a "stand alone." In addition, biomarkers for both efficacy and toxicity that would be accessible in humans are being tested and validated and GLP studies will soon be conducted. Following successful completion of these studies, we will file an IND and initiate a Phase I clinical trial.